物件追蹤是電腦視覺中最重要和最快速發展的領域。另一方面，360°影片和其相關技術在這些年來越來越受歡迎。因此在本論文中，我們提出了一種基於視圖翹曲的方法來增強和擴展現有的在線物件追蹤算法，使其得以用於追蹤360°影片。此外，我們還提出了一個新的標準數據集，用於評估360°影片物件追蹤算法和分析追蹤結果。

Object tracking is one of the most important and rapidly developing fields in computer vision. On the other hand, 360° videos and related technologies are getting popular these years. In this thesis, we present a view-warping based method to enhance and expand existing online object tracking algorithms to be used on 360° videos. In addition, we propose a new benchmark dataset for evaluating 360° video object tracking algorithms and for analyzing the tracking results.

1 Introduction 11
1.1 Object Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 360 Video Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2 Related Work 15
2.1 Object Tracking Methods . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2 Baseline: ECO tracker . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Baseline: fDSST tracker . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 Our Approach 20
3.1 View-warping Method . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2 Implementation based on baselines . . . . . . . . . . . . . . . . . . . 23
4 Sports-360 Tracking Dataset 25
5 Experiments 30
5.1 Implementation Details . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6 Conclusions and Future Work 37

[1] P. H. S. T. Afshin Dehghan, Yicong Tian and M. Shah. Target identity-aware
network
ow for online multiple target tracking. In CVPR, 2015.
[2] W. Choi. Near-online multi-target tracking with aggregated local
ow descriptor.
CoRR, abs/1504.02340, 2015.
[3] M. Danelljan, G. Bhat, F. S. Khan, and M. Felsberg. ECO: ecient convolution
operators for tracking. CoRR, abs/1611.09224, 2016.
[4] M. Danelljan, G. Hager, F. S. Khan, and M. Felsberg. Discriminative scale space
tracking. IEEE Trans. Pattern Anal. Mach. Intell., 39(8):1561{1575, 2017.
[5] M. Danelljan, A. Robinson, F. S. Khan, and M. Felsberg. Beyond correlation
lters: Learning continuous convolution operators for visual tracking. CoRR,
abs/1608.03773, 2016.
[6] A. Delforouzi, S. A. H. Tabatabaei, K. Shirahama, and M. Grzegorzek. Polar
object tracking in 360-degree camera images. In IEEE International Symposium
on Multimedia (ISM 2016), pages 347{352, December 2016.
[7] Facebook. Facebook surround 360. https://facebook360.fb.com/
facebook-surround-360/.
[8] H. Hu, Y. Lin, M. Liu, H. Cheng, Y. Chang, and M. Sun. Deep 360 pilot: Learning
a deep agent for piloting through 360 sports video. CoRR, abs/1705.01759,
2017.
[9] H. Kieritz, S. Becker, W. Hubner, and M. Arens. Online multi-person tracking
using integral channel features. 2016 13th IEEE International Conference on
Advanced Video and Signal Based Surveillance (AVSS), 00:122{130, 2016.
37
[10] M. Kristan, A. Leonardis, J. Matas, M. Felsberg, R. P
ugfelder, L. Cehovin,
T. Vojir, G. Hager, A. Lukezic, and G. Fernandez. The visual object tracking
vot2016 challenge results. Springer, Oct 2016.
[11] M. Kristan, R. P
ugfelder, A. Leonardis, J. Matas, L. Cehovin, G. Nebehay,
T. Vojir, G. Fernandez, A. Lukezic, A. Dimitriev, A. Petrosino, A. Saari, B. Li,
B. Han, C. Heng, C. Garcia, D. Pangersic, G. Hager, F. S. Khan, F. Oven,
H. Possegger, H. Bischof, H. Nam, J. Zhu, J. Li, J. Y. Choi, J.-W. Choi, J. F.
Henriques, J. van de Weijer, J. Batista, K. Lebeda, K. Ofjall, K. M. Yi, L. Qin,
L. Wen, M. E. Maresca, M. Danelljan, M. Felsberg, M.-M. Cheng, P. Torr,
Q. Huang, R. Bowden, S. Hare, S. Y. Lim, S. Hong, S. Liao, S. Hadeld, S. Z.
Li, S. Duner, S. Golodetz, T. Mauthner, V. Vineet, W. Lin, Y. Li, Y. Qi, Z. Lei,
and Z. Niu. The visual object tracking vot2014 challenge results. In L. Agapito,
M. M. Bronstein, and C. Rother, editors, Proceedings, European Conference on
Computer Vision (ECCV) Visual Object Tracking Challenge Workshop, volume
8926 of Lecture Notes in Computer Science, pages 191{217, Zurich, Switzerland,
September 2014. Springer International Publishing.
[12] F. Perazzi, J. Pont-Tuset, B. McWilliams, L. Van Gool, M. Gross, and
A. Sorkine-Hornung. A benchmark dataset and evaluation methodology for video
object segmentation. In Computer Vision and Pattern Recognition, 2016.
[13] A. Roshan Zamir, A. Dehghan, and M. Shah. Gmcp-tracker: Global multiobject
tracking using generalized minimum clique graphs. In Proceedings of the
European Conference on Computer Vision (ECCV), 2012.
[14] S. Tang, M. Andriluka, B. Andres, and B. Schiele. Multi people tracking with
lifted multicut and person re-identication. In IEEE Conf. on Computer Vision
and Pattern Recognition (CVPR), July 2017.
[15] Y. Wu, J. Lim, and M.-H. Yang. Online object tracking: A benchmark. In IEEE
Conference on Computer Vision and Pattern Recognition (CVPR), 2013.
[16] Y.Wu, J. Lim, and M.-H. Yang. Object tracking benchmark. IEEE Transactions
on Pattern Analysis & Machine Intelligence, 37(9):1834{1848, 2015.
38
[17] Y. Xiang, A. Alahi, and S. Savarese. Learning to track: Online multi-object
tracking by decision making. In International Conference on Computer Vision,
2015.